GL_RGB32F is available only when texture_buffer_object_rgb32 is supported.
As the test covers different topic simply replace the format with one
that is supported by both core and compat profile and does not require
any extensions.
Cc: Jacob Penner
Signed-off-by: Emil Velikov
---
.../spec/gl
GL_RGB32F is available only when texture_buffer_object_rgb32 is supported.
As the test covers different topic simply replace the format with one
that is supported by both core and compat profile and does not require
any extensions.
Cc: Jacob Penner
Signed-off-by: Emil Velikov
---
.../spec/gl
Cc: Fabian Bieler
Signed-off-by: Emil Velikov
---
tests/spec/arb_clear_buffer_object/formats.c | 22 ++
1 file changed, 10 insertions(+), 12 deletions(-)
diff --git a/tests/spec/arb_clear_buffer_object/formats.c
b/tests/spec/arb_clear_buffer_object/formats.c
index fb516bb
On 04/03/14 20:42, Anuj Phogat wrote:
> On Tue, Mar 4, 2014 at 9:18 AM, Emil Velikov wrote:
>> GL_RGB32F is available only when texture_buffer_object_rgb32 is supported.
>> As the test covers different topic simply replace the format with one
>> that is supported by both co
ncentive to forget them.
The buggers in arb_clear_buffer_object/formats.c are my fault :'(
FWIW
Reviewed-by: Emil Velikov
> tests/glx/glx-buffer-age.c | 2 +-
> tests/glx/glx-copy-sub-buffer.c
On 17/06/14 23:50, Ian Romanick wrote:
> On 06/16/2014 11:29 PM, Ilia Mirkin wrote:
>> All the arb_shader_atomic_counters tests start with
>>
>> config.supports_gl_core_version = 31;
>>
>> However when I try to run them on the NVIDIA proprietary driver, I get:
>>
>> piglit: info: Requested
Hello list,
The series adds a piglit_wgl_framework based on my "Add WGL support
to waffle" GSoC this summer. A sizeable chunk of the waffle code is not
yet upstream, although I would expect that to change soon :P
All but patches 04, 10 and 11 should be safe to land even without waffle
in place
ning individual tests, otherwise
the test will abort after being displayed for 8 seconds.
TODO:
- Bump the version requirement, once a WAFFLE_WGL is released.
- Update the instructions in the README.
- Add input handling (event_loop).
Signed-off-by: Emil Velikov
---
One does not need to build either one if the target does not
have/support them. The upcoming waffle WGL support is a nice example.
Signed-off-by: Emil Velikov
---
tests/util/CMakeLists.txt | 12 ++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/tests/util
7;bool' is
illegal.
Signed-off-by: Emil Velikov
---
include/msvc/c99/stdbool.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/msvc/c99/stdbool.h b/include/msvc/c99/stdbool.h
index dd55255..089f8d4 100644
--- a/include/msvc/c99/stdbool.h
+++ b/include/msvc/c99/s
The function is part of the secapi, which is not available
under WinXP. Building piglit with it will result in broken
tests.
Signed-off-by: Emil Velikov
---
CMakeLists.txt | 2 ++
1 file changed, 2 insertions(+)
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 36b06fc..5807f63 100644
--- a
- Mandating pkg-check when building with MSVC is silly.
- Do not mess around with CMAKE_C*_FLAGS directly but use
include_directories to handle waffle's headers location.
- Use WAFFLE_LIBRARIES over WAFFLE_LDFLAGS for linking purposes.
Signed-off-by: Emil Velikov
---
CMakeList
I'm not entirely sure how piglit build with gcc as is, yet VC compiler
seems very unhappy about this.
Signed-off-by: Emil Velikov
---
tests/util/piglit-vbo.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/util/piglit-vbo.cpp b/tests/util/piglit-vbo.cpp
index 8a
piglit-dispatch.c is about to get hairy even without this hunk. Also
moving the code in piglit-dispatch-init.c will allow us to have a clear
visual as we start removing the non-waffle (glut) support.
Signed-off-by: Emil Velikov
---
tests/util/piglit-dispatch-init.c | 85
Avoid building the code for platforms that lack wayland. The note is
still valid as it is based on the lack of input handling (event_loop)
when building with wayland in mind. To avoid issues as that changes just
wrap it up now, similar to every other platform.
Signed-off-by: Emil Velikov
Already declared in tests/util/piglit-framework-gl.h.
Signed-off-by: Emil Velikov
---
tests/util/sized-internalformats.h | 3 ---
1 file changed, 3 deletions(-)
diff --git a/tests/util/sized-internalformats.h
b/tests/util/sized-internalformats.h
index 9529d7c..e595b07 100644
--- a/tests/util
nd hiding it at window_destroy will trigger an
assert.
Signed-off-by: Emil Velikov
---
tests/util/piglit-framework-gl/piglit_wfl_framework.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/tests/util/piglit-framework-gl/piglit_wfl_framework.c
b/tests/util/pigl
For some bizzare reason the MSVC build fails with unresolved symbols
without these two set in. I'm suspecting that something fishy is
happening with waffle and/or cmake + MSVC. Until we have that one
resolved, this helps us get through with the build.
Signed-off-by: Emil Velikov
---
tests
Signed-off-by: Emil Velikov
---
tests/util/piglit-framework-gl/piglit_winsys_framework.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/tests/util/piglit-framework-gl/piglit_winsys_framework.c
b/tests/util/piglit-framework-gl/piglit_winsys_framework.c
index 5d7dae1..c80e972 100644
--- a
Already declared in tests/util/piglit-framework-gl.h.
v2: Also remove piglit_init(). Spotted by Ilia.
Signed-off-by: Emil Velikov
---
Thanks Ilia. I was casing rather nasty bug and did not notice
piglit_init().
-Emil
tests/util/sized-internalformats.h | 6 --
1 file changed, 6
On 12/08/14 17:30, Ilia Mirkin wrote:
> On Tue, Aug 12, 2014 at 1:18 PM, Emil Velikov
> wrote:
>> I'm not entirely sure how piglit build with gcc as is, yet VC compiler
>> seems very unhappy about this.
>
> http://www.open-std.org/jtc1/sc22/wg14/www/docs/n11
On 12/08/14 22:21, Ilia Mirkin wrote:
> On Tue, Aug 12, 2014 at 5:19 PM, Ilia Mirkin wrote:
>> On Tue, Aug 12, 2014 at 5:18 PM, Emil Velikov
>> wrote:
>>> On 12/08/14 17:30, Ilia Mirkin wrote:
>>>> On Tue, Aug 12, 2014 at 1:18 PM, Emil Velikov
>>
Hello list,
Recently I've been trying out piglit under Windows, and by default the cmd
window is 80 columns in width. As such as soon as the status line becomes
larger than 80 characters, a separate new line is printed for each test
iteration. AFAICS the issue exist under Linux as well, yet the de
On 12/08/14 23:10, Dylan Baker wrote:
> On Tuesday, August 12, 2014 11:01:34 PM Emil Velikov wrote:
>> Hello list,
>>
>> Recently I've been trying out piglit under Windows, and by default the
> cmd
>> window is 80 columns in width. As such as soon as the statu
On 13/08/14 21:59, Dylan Baker wrote:
> On Wednesday, August 13, 2014 02:16:23 PM Emil Velikov wrote:
>> On 12/08/14 23:10, Dylan Baker wrote:
>>> On Tuesday, August 12, 2014 11:01:34 PM Emil Velikov wrote:
>>>> Hello list,
>>>>
>>>> Rece
On 17/08/14 19:03, Ilia Mirkin wrote:
> The test relies on level 1 being used from the texture when rendering to
> the window.
>
AFAICS Brian's recent "set minimum windows dimentions to AxB" was inspired by
a Windows feature - if a window has a titlebar (and/or assositated buttons)
it's minimum si
On 13/08/14 14:08, Brian Paul wrote:
> On 08/12/2014 11:18 AM, Emil Velikov wrote:
>> Hello list,
>>
>> The series adds a piglit_wgl_framework based on my "Add WGL support
>> to waffle" GSoC this summer. A sizeable chunk of the waffle code is not
>> ye
On 18/08/14 15:22, Jose Fonseca wrote:
> On 12/08/14 18:18, Emil Velikov wrote:
> [...]
>> - Use WAFFLE_LIBRARIES over WAFFLE_LDFLAGS for linking purposes.
> [...]
>
> Emil,
>
> This particular change is causing the build to fail when libwaffle-1.so is not
> on a
On 18/08/14 16:49, jfons...@vmware.com wrote:
> From: José Fonseca
>
Had a chat with the guys at #cmake and it seems like I've got it wrong - one
ought to use *LDFLAGS, esp. if the library is installed outside of /usr/lib.
Reviewed-by: Emil Velikov
> Otherwise the build fails wh
Hello list,
When generating HTML summary of the results some symbols are not escaped
correctly, thus the links created for the related tests are invalid. See the
following examples
glsl1-Preprocessor test 11 (#elif)
glean/glsl1-Preprocessor%20test%2011%20(#elif).html
glsl1-Preprocessor test 15 (
On 03/09/14 07:14, Dylan Baker wrote:
> CC: Emil Velikov
> Signed-off-by: Dylan Baker
Works like a charm, thank you :)
Tested-by: Emil Velikov
> ---
> framework/summary.py | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/framework/summary.py b
ate arguments changed a few of the names to
>>> shorter versions, including environment to env. The template was updated
>>> incorrectly in the template, and environment has not been printed since
>>
>> The template was updated incorrectly in the template?
>
>
&g
Hi Brian,
Wouldn't it be better if we add a couple of wrapper functions: piglit_get_time
and piglit_sleep ? This way we'll be able to keep the tests clean and
OS-agnostic and avoid the trivial duplication introduced with the second patch
:)
Cheers,
Emil
On 04/09/14 20:34, Brian Paul wrote:
> Fr
On 04/09/14 21:56, Brian Paul wrote:
> On 09/04/2014 02:25 PM, Emil Velikov wrote:
>> Hi Brian,
>>
>> Wouldn't it be better if we add a couple of wrapper functions:
>> piglit_get_time
>> and piglit_sleep ? This way we'll be able to keep the tests cle
On 04/09/14 21:54, Dylan Baker wrote:
> This solves the bug of running piglit run -n 'wip/foo' which causes a
> number of issues in the html summary generation.
>
>From a quick look it seems that it might help with a funny issue that I'm
>seeing:
Whenever piglit is ran on Windows the testname us
On 04/09/14 23:18, Brian Paul wrote:
> On 09/04/2014 03:07 PM, Emil Velikov wrote:
>> On 04/09/14 21:56, Brian Paul wrote:
>>> On 09/04/2014 02:25 PM, Emil Velikov wrote:
>>>> Hi Brian,
>>>>
>>>> Wouldn't it be better if we add a
On 04/09/14 22:55, Dylan Baker wrote:
> On Thursday, September 04, 2014 10:49:10 PM Emil Velikov wrote:
>> On 04/09/14 21:54, Dylan Baker wrote:
>>> This solves the bug of running piglit run -n 'wip/foo' which causes a
>>> number of issues in the html summary
On 25/09/14 15:34, Brian Paul wrote:
> On 09/25/2014 02:32 AM, Juliet Fru wrote:
>>
>> Hello Brian,
>>
>>
>> There's definitely work that can be done for Piglit. For example,
>> porting the old Glean tests to piglit's framework.
>>
>>
>> I am currently downloading the code and would build
Latest waffle (1.4.0) requires cmake 2.8.11 and considering that
waffle and piglit are build hand-in-hand, do the same here.
Suggested-by: Chad Versace
Signed-off-by: Emil Velikov
---
I'm not entirely sure how many people build waffle separate
of piglit and if we must bump the requirement
Hello all,
As you already know I've been doing mesa releases for a bit now, and one
of the things that I found "inspiring" is the lack of piglit releases.
Take the following
"I've tested Mesa X and there are no regression... well there are but
that's because I unintentionally updated piglit to 5c
fine with me.
>
> I'll add that Gentoo has 2.8.12 as stable, and 3.0.2 as unstable
>
> On Friday, September 26, 2014 11:50:53 AM Emil Velikov wrote:
>> Latest waffle (1.4.0) requires cmake 2.8.11 and considering that
>> waffle and piglit are build hand-in-hand, do the sa
fine with me.
>
> I'll add that Gentoo has 2.8.12 as stable, and 3.0.2 as unstable
>
> On Friday, September 26, 2014 11:50:53 AM Emil Velikov wrote:
>> Latest waffle (1.4.0) requires cmake 2.8.11 and considering that
>> waffle and piglit are build hand-in-hand, do the sa
On 30/09/14 16:18, Ian Romanick wrote:
> On 09/29/2014 10:01 AM, Matt Turner wrote:
>> On Mon, Sep 29, 2014 at 9:34 AM, Emil Velikov
>> wrote:
>>> While I've been through the RELEASES document I believe it would be
>>> beneficial if we regularly create
So in a nutshell what I've gathered so far:
* Concerns that it will take too much effort.
Dare I say it - it will not. A single build + sanity run takes a few
minutes.
* There is no point if tagging/shipping (distro or otherwise) piglit
for people that do not know/have the time/etc to build it.
Reported-by: Ilia Mirkin
Signed-off-by: Emil Velikov
---
Untested, but if my understanding is correct, it should handle
most cases.
-Emil
tests/util/piglit-framework-gl/piglit_winsys_framework.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/tests/util/piglit-framework-gl
Otherwise we end up leaking the drawable. Spotted by Valgrind
Signed-off-by: Emil Velikov
---
Perhaps we can simplify the call chain for the following a bit
make_current > make_current_singlepass > special_case31 >
make_curren_singlepass...
But that for another day :P
-Emil
t
On 02/10/14 06:21, Dylan Baker wrote:
> Should the subject say 'unbind' rather than 'unbound'?
>
Indeed it should. I had the feeling that it does not sounds quite right,
thanks for spotting :)
Planning to get this an the other leakfix (incl. Ilia's t-b, and the
typo/grammar fixed) over the weeken
On 02/10/14 06:21, Dylan Baker wrote:
> Should the subject say 'unbind' rather than 'unbound'?
>
Indeed it should. I had the feeling that it does not sounds quite right,
thanks for spotting :)
Planning to get this an the other leakfix (incl. Ilia's t-b, and the
typo/grammar fixed) over the weeken
On 02/10/14 21:44, Ian Romanick wrote:
> On 09/30/2014 09:55 AM, Emil Velikov wrote:
>> On 30/09/14 16:18, Ian Romanick wrote:
>>> On 09/29/2014 10:01 AM, Matt Turner wrote:
>>>> On Mon, Sep 29, 2014 at 9:34 AM, Emil Velikov
>>>> wrote:
>>>
On 02/10/14 21:46, Ian Romanick wrote:
> On 09/30/2014 10:05 AM, Emil Velikov wrote:
>> So in a nutshell what I've gathered so far:
>>
>> * Concerns that it will take too much effort.
>> Dare I say it - it will not. A single build + sanity run takes a few
>>
Hello all,
Have anyone noticed an interesting new "feature" in the html info page
of the respective piglit run. It seems that now uname, lspci and glxinfo
are merged with the options.
I would assume that this is not intentional, but a pair of parents may
be missing somewhere ?
Thanks,
Emil
__
Hi Dylan,
On 04/10/14 01:57, Dylan Baker wrote:
> We have recently started using jenkins here at Intel to do regression
> testing across multiple generations of hardware. Because of this we have
> become concerned with the large number of tests that fail and the large
> number of tests that skip,
gt; This patch corrects that problem, making the final output look much
> nicer.
>
Nice one Dylan, I've spotted this one but did not really bother with
reporting it :P
Tested-by: Emil Velikov
> Signed-off-by: Dylan Baker
___
Piglit mail
to be more robust.
Reported-and-Tested-by: Emil Velikov
Slightly sucks that I would need to manually edit the faulty result
files, but it's way better than rerunning everything from scratch
again :P
Thank you Dylan.
Emil
___
Pig
On 30 October 2014 21:27, Dylan Baker wrote:
> This corrects in a permanent and automatic way the changes in the
> results that were fixed in the previous patch.
>
> v2: - add this patch
>
> Signed-off-by: Dylan Baker
> ---
>
> Emil, this should solve the manually editing files problem you report
t;
Works like a charm. Thanks.
Tested-by: Emil Velikov
> Signed-off-by: Dylan Baker
___
Piglit mailing list
Piglit@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/piglit
On 07/11/14 14:21, jfons...@vmware.com wrote:
> From: José Fonseca
>
> As the semantics of sprintf_s's arguments are quite different, so code
> that relies on it might misbehave.
Indeed sprintf_s has an additional argument (buffer_length) which might
cause a bit of an issue.
On a slightly relate
)
> +Start Menu->All Programs->Visual Studio 2013->Visual Studio Tools->VS2012
> x86 Native Tools Command Prompt
Have a strange feeling that the menu says "VS2013 x86..."
Either way the series looks good.
FWIW Reviewed-by: Emil Velikov
Cheers,
Emil
On 07/11/14 14:21, jfons...@vmware.com wrote:
> From: José Fonseca
>
> As these are not supported on MSVC, not even MSVC 2013 since it only has
> library support for C99, it doesn't actually support C99 syntax.
>
I do recall that msvc2013 just errors out in such cases, yet I've not
seen any issu
On 14/11/14 14:57, jfons...@vmware.com wrote:
> From: José Fonseca
>
> https://www.opengl.org/sdk/docs/man2/xhtml/glXChooseFBConfig.xml states
> that
>
> "GLX_DOUBLEBUFFER Must be followed by True or False."
>
> No idea how this ever passed given the buffer overflow that ensued.
>
I've recent
On 18/11/14 15:14, Brian Paul wrote:
> On 11/17/2014 05:00 PM, Ben Widawsky wrote:
>> Signed-off-by: Ben Widawsky
[snip]
>> +static struct query queries[] = {
>> +{
>> + .query = GL_GEOMETRY_SHADER_INVOCATIONS,
>> + .name = "GL_GEOMETRY_SHADER_INVOCATIONS",
>> + .expected = NUM_PRI
On 18/11/14 18:25, Jose Fonseca wrote:
> You're sure? I thought that MSVC 2013 only had _library_ support for C99, but
> I confess I haven't tested..
>
Note that I've mentioned named/designated initialisers rather than the
whole C99. And yes building executables with named initializers works.
Ch
Signed-off-by: Emil Velikov
---
tests/util/piglit-glx-util.c | 7 ---
1 file changed, 7 deletions(-)
diff --git a/tests/util/piglit-glx-util.c b/tests/util/piglit-glx-util.c
index 5e3152f..3042dc8 100644
--- a/tests/util/piglit-glx-util.c
+++ b/tests/util/piglit-glx-util.c
@@ -498,10 +498,3
Hello list,
This is my (hopefully) final submission on the topic of converting
piglit to use Waffle WGL. With some cleanups (input handling?) to
follow some time later.
Since the original submission we have dropped the Waffle changes
which break the API, and WGL support is now in waffle/master :
Signed-off-by: Emil Velikov
---
tests/util/piglit-util.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/util/piglit-util.h b/tests/util/piglit-util.h
index 84e2066..9731c1c 100755
--- a/tests/util/piglit-util.h
+++ b/tests/util/piglit-util.h
@@ -115,7 +115,7 @@ enum
le (no API breakage)
Signed-off-by: Emil Velikov
Reviewed-by: Brian Paul (v1)
Signed-off-by: Emil Velikov
---
CMakeLists.txt | 7 +-
README | 38 +-
tests/util/CMakeLists.txt | 5 ++
piglit-dispatch.c is about to get hairy even without this hunk. Also
moving the code in piglit-dispatch-init.c will allow us to have a clear
visual as we start removing the non-waffle (glut) support.
Signed-off-by: Emil Velikov
Reviewed-by: Brian Paul
Signed-off-by: Emil Velikov
---
tests
ndows
result file.
Cc: Dylan Baker
Signed-off-by: Emil Velikov
---
framework/summary.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/framework/summary.py b/framework/summary.py
index f4fd80d..6dc2b07 100644
--- a/framework/summary.py
+++ b/framework/summary.py
@@ -482,7 +
ndows
result file.
v2: Attempt to handle Linux results on Windows.
Cc: Dylan Baker
Signed-off-by: Emil Velikov
---
framework/summary.py | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/framework/summary.py b/framework/summary.py
index f4fd80d..1bb5297 100644
--- a/fram
against them, but located
in a different folder, see the next todo note.
- Modify the python scripts to set PATH to the location of the DLLs.
Alternatively apply any of the other solutions as mentioned here
http://msdn.microsoft.com/en-us/library/7d83bc18.aspxy
Cc: José Fonseca
Signed-off-by: Emil
libglut, which explicitly pulls libGL,
thus glXGetProcAddress{,ARB} ends up resolved.
Signed-off-by: Emil Velikov
---
tests/util/CMakeLists.txt | 4
1 file changed, 4 insertions(+)
diff --git a/tests/util/CMakeLists.txt b/tests/util/CMakeLists.txt
index 98eedd0..a6ae6dd 100644
--- a/tests/util/
Hi all,
Feeling inspired by Jose's recent work I decided to do a bit of
cleanup :)
The series aims to:
- Nuke/move the remaining ifdef _WIN32 _MSC_VER hacks into tests/util.
- Improve piglit_getmicroseconds (add fallback for CLOCK_MONOTONIC)
- Convert {arb,ext}_timer_query to use it. On pape
Signed-off-by: Emil Velikov
---
tests/spec/arb_timer_query/timestamp-get.c | 6 +-
tests/util/piglit-util.h | 4
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/tests/spec/arb_timer_query/timestamp-get.c
b/tests/spec/arb_timer_query/timestamp-get.c
from shader_runner.
Signed-off-by: Emil Velikov
---
tests/shaders/shader_runner.c | 6 --
1 file changed, 6 deletions(-)
diff --git a/tests/shaders/shader_runner.c b/tests/shaders/shader_runner.c
index 628ef44..107e2b5 100644
--- a/tests/shaders/shader_runner.c
+++ b/tests/shaders/shader_run
Fallback to gettimeofday, if clock_gettime supports monotonic yet
fails to get the time.
v2: Fix gettimeofday fallback.
Signed-off-by: Emil Velikov
---
tests/util/piglit-util.c | 19 +++
1 file changed, 15 insertions(+), 4 deletions(-)
diff --git a/tests/util/piglit-util.c b
The above uses clock_gettime(CLOCK_MONOTONIC) (when available) rather
than gettimeofday(). The former has greather precision and could help
with the inconsintent results that people are getting.
Signed-off-by: Emil Velikov
---
tests/spec/arb_timer_query/timestamp-get.c | 31
Move the final two windows specific #undef next to their min/max
brothers. Note that they seem to be defined when building with MSVC
(_MSV_VER) rather than all windows platfroms (_WIN32).
Signed-off-by: Emil Velikov
---
tests/general/fog-modes.c| 5 -
tests/spec/ext_fog_coord
This way we can actually reuse it in {arb,ext}_timer_query
Signed-off-by: Emil Velikov
---
tests/spec/glx_oml_sync_control/timing.c | 8
tests/texturing/teximage-colors.c| 6 +++---
tests/util/piglit-util.c | 4 ++--
tests/util/piglit-util.h | 6
Rather than assuming that piglit_get_microseconds() returns -1 when
there is no monotonic timer, add explicit function to check.
Use it where needed, and break the above assumption.
Signed-off-by: Emil Velikov
---
tests/spec/glx_oml_sync_control/timing.c | 2 +-
tests/util/piglit-util.c
Signed-off-by: Emil Velikov
---
tests/shaders/shader_runner.c | 27 ---
tests/util/piglit-util.h | 27 +++
2 files changed, 27 insertions(+), 27 deletions(-)
diff --git a/tests/shaders/shader_runner.c b/tests/shaders/shader_runner.c
index
The above uses clock_gettime(CLOCK_MONOTONIC) (when available) rather
than gettimeofday(). The former has greather precision and could help
with the inconsintent results that people are getting.
Signed-off-by: Emil Velikov
---
tests/spec/ext_timer_query/time-elapsed.c | 60
Signed-off-by: Emil Velikov
---
tests/util/piglit-util-cl-enum.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/tests/util/piglit-util-cl-enum.c b/tests/util/piglit-util-cl-enum.c
index 9076387..5a38e01 100644
--- a/tests/util/piglit-util-cl-enum.c
+++ b/tests/util/piglit-util-cl-enum.c
Typecast malloc() return value, of the CPP compiler will hate us.
Signed-off-by: Emil Velikov
---
tests/util/piglit-util.c | 27 ---
tests/util/piglit-util.h | 36
2 files changed, 28 insertions(+), 35 deletions(-)
diff --git a/tests
This commit effectively reverts the following commit
0f0543fd1b2(Portability build fixes for MinGW.)
Confirmed that building with mingw-w64 works without a problem.
Cc: José Fonseca
Signed-off-by: Emil Velikov
---
tests/shaders/glsl-fs-pointcoord.c | 4
1 file changed, 4 deletions
On 23/11/14 09:38, Jose Fonseca wrote:
> On 21/11/14 19:52, Emil Velikov wrote:
[snip]
>> --- /dev/null
>> +++ b/tests/util/piglit-framework-gl/piglit_wgl_framework.c
[snip]
>> +static void
>> +enter_event_loop(struct piglit_winsys_framework *winsys_fw)
>> +{
>
On 23/11/14 09:46, Jose Fonseca wrote:
> On 22/11/14 19:36, Emil Velikov wrote:
>> Seemingly this is sufficient for a Windows piglit run/result file to be
>> ran under Linux and the correct directory structure (and html) to be
>> generated.
>>
>> Unfortunately m
On 23/11/14 11:18, Jose Fonseca wrote:
> On 22/11/14 22:26, Emil Velikov wrote:
>> Rather than rebuilding every single test as we change the util
>> libraries and increase the size of each test by ~9MiB (as noticed
>> in the mingw-w64 build), just revert to shared piglituti
Afaict in some cases we don't use posixpath, as "key" ends up with a mix
of both \ and /. Any ideas what/where to look to fix this properly ?
-Emil
On 23/11/14 17:54, Dylan Baker wrote:
> Isn't this why we were using posixpath explicitly?
> On Nov 23, 2014 7:15
On 23/11/14 20:14, Jose Fonseca wrote:
> On 23/11/14 18:11, Emil Velikov wrote:
>> On 23/11/14 11:18, Jose Fonseca wrote:
>>> On 22/11/14 22:26, Emil Velikov wrote:
>>>> Rather than rebuilding every single test as we change the util
>>>> libraries and i
Similar to the earlier patch for waffle-less (glut) builds. The need
for this patch became apparent as we removed the libGL link dependency
in waffle.
Cc: Mark Janes
Reported-by: Mark Janes
Signed-off-by: Emil Velikov
---
tests/util/CMakeLists.gl.txt | 4
1 file changed, 4 insertions
Similar to the earlier patch for waffle-less (glut) builds. The need
for this patch became apparent as we removed the libGL link dependency
in waffle.
Cc: Mark Janes
Reported-by: Mark Janes
Signed-off-by: Emil Velikov
---
tests/util/CMakeLists.gl.txt | 4
1 file changed, 4 insertions
On 25/11/14 18:45, Mark Janes wrote:
> This patched fixed the gl tests, but I encountered the same error for
> gles1/gles2/gles3.
>
> Piglit built for me when I added the same link instruction to:
>
> tests/util/CMakeLists.gles1.txt
> tests/util/CMakeLists.gles2.txt
> tests/util/CMakeLists.gles3.
Similar to the earlier patch for waffle-less (glut) builds. The need
for this patch became apparent as we removed the libGL link dependency
in waffle.
v2: Try to handle both gl and gles*.
Cc: Mark Janes
Reported-by: Mark Janes
Signed-off-by: Emil Velikov
---
tests/util/CMakeLists.txt | 5
On 01/12/14 18:17, Chad Versace wrote:
>
>
> On 12/01/2014 09:41 AM, Chad Versace wrote:
>> On 11/26/2014 01:11 AM, Emil Velikov wrote:
>>
>>> Things are not as black and white as you might think. Most of piglit
>>> uses what I call "target
On 01/12/14 18:17, Chad Versace wrote:
> On 11/26/2014 01:11 AM, Emil Velikov wrote:
>> Similar to the earlier patch for waffle-less (glut) builds. The need
>> for this patch became apparent as we removed the libGL link dependency
>> in waffle.
>>
>> v2: Try to h
The test itself is written against the ARB extension, and neither the
test or the ARB extension requires the EXT one.
Signed-off-by: Emil Velikov
---
It's unlikely that anyone will hit this (i.e. has support for ARB but
lacks the EXT extension) but from a quick look at the spec it seems
By the time piglit_init() is executed, the context is creates and the dispatch
table has been initialised.
Signed-off-by: Emil Velikov
---
tests/texturing/getteximage-luminance.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/texturing/getteximage-luminance.c
b/tests
On 03/12/14 14:44, Brian Paul wrote:
> On 12/03/2014 06:37 AM, Emil Velikov wrote:
>> The test itself is written against the ARB extension, and neither the
>> test or the ARB extension requires the EXT one.
>>
>> Signed-off-by: Emil Velikov
>> ---
>>
>
On 03/12/14 15:37, Jose Fonseca wrote:
> From: José Fonseca
>
> Now that this is an inline function, the warning appears all over the
> place.
> ---
> tests/util/piglit-util.h | 5 +
> 1 file changed, 5 insertions(+)
>
> diff --git a/tests/util/piglit-util.h b/tests/util/piglit-util.h
> ind
Thanks for the quick fixes José.
Apart from a nitpick in patch 2 the series looks good.
-Emil
On 03/12/14 15:37, Jose Fonseca wrote:
> From: José Fonseca
>
> And not just MSVC.
>
> Fixes build with MinGW.
> ---
> tests/util/piglit-util.h | 10 ++
> 1 file changed, 6 insertions(+), 4
1 - 100 of 329 matches
Mail list logo
